In dendrites, Ca2+ is critical in determining how neurons respond to incoming excitation. While numerous studies have focused on how dendritic Ca2+ relates to behaviorally-relevant neuronal and circuit activity using correlative observations, there is currently no method to precisely manipulate Ca2+ in neurons in vivo and thus causally test its role in circuit function and behavior. In non-neuronal cells, mitochondria can act as sinks for Ca2+ released from the endoplasmic reticulum (ER) by forming direct contacts with these concentrated intracellular Ca2+ stores. Recently the Polleux lab discovered that protein PDZD8 enables mitochondria to buffer ER-released Ca2+ in dendrites by tethering these organelles together; in the absence of PDZD8, cytosolic [Ca2+] is markedly higher after synaptically-evoked ER release. Using a newly-developed Pdzd8 conditional knockout (cKO) mouse line, versatile recombination and labeling strategies, and a newly-developed optogenetic tool to rapidly and reversibly induce new ER-mitochondria contacts with light, we are now poised to directly manipulate the spatial and temporal dynamics of dendritic Ca2+ in awake and behaving mice. By combining these approaches with 2-photon Ca2+ imaging during head- fixed behavior, we will causally test the relationship between dendritic Ca2+ dynamics and neuronal input-output transformations, circuit function, and learning & memory. Using hippocampal CA1 pyramidal neurons (PNs) as a model system, we will further assess these relationships with respect to input-specific dendritic compartments thought to receive distinct streams of behaviorally-relevant information. The long-term objective of this proposal is to create a platform for systematically and quantitatively probing the transformation of subcellular Ca2+ dynamics into higher-order cognitive processes in health and disease. While the current proposal seeks to establish this novel platform in CA1 PNs in the context of spatial learning, we aim for general applicability to the study of subcellular Ca2+ dynamics in higher-order brain processes. Hypothesis: We hypothesize that dendritic Ca2+ is integrated in an input-specific manner in CA1 PN apical dendrites to drive circuit dynamics underlying spatial learning and memory. We will test this hypothesis in the following specific aims: Specific Aim 1: Characterize ER-mitochondria tethering as a novel inroad to bidirectionally manipulating Ca2+ dynamics in input-defined dendritic compartments of CA1 PNs. Specific Aim 2: Causally test the link from dendritic Ca2+ dynamics in CA1 PNs to circuit-level neural activity and spatial learning in vivo.